100 g of coarsely powdered ephedra are added to a solution of 50 g sodium carbonate in 150 ml water. Liquid and powder are thoroughly mixed by means of a glass rod in a 1000 ml beaker and left there sitting for 2 hours.
Then follows the percolation in a percolator, but a Hempel column with a cork and a stopcock is a perfect substitute for this !
Percolator in the lower end (just in front of the stopcock) is closed with a pad of glass wool, on which rests a bed of sea sand about 10 mm thick. Percolation tube is then filled to about 1/3 of its length with dichloromethane. The crumbly paste obtained from ephedra and sodium carbonate solution is then transferred into the percolator (stirring with glass rod is necessary to drive out air bubbles). The drug is tamped with glass wool and continously percolated with 1000 ml dichloromethane.
The percolate is rotovapped to a volume of about 5 ml. To the so concentrated extract 30 ml of ether are added and the ethereal solution is extracted in a sep funnel with 3x 50 ml 0.5 N hydrochloric acid. The acidic extracts are pooled and are filtrated over 50 g potassium carbonate in a funnel equipped with a fluted filter paper. The now neutral to slightly alkaline solution is then extracted with 4x 75 ml dichloromethane. The extracts are pooled and dried over anhydrous sodium sulfate.
The dichloromethane is distilled off and the residue, a yellow-orange oil, is dissolved in a mixture of 20 ml of ether and 20 ml of dichloromethane. The solution is then gassed with HCl, made from 20 ml of concentrated sulfuric acid and 10 g sodium chloride. The gas is dried by passing through a wash bottle filled with 50 ml of concentrated sulfuric acid. Excess of HCl is absorbed in 10 % NaOH-solution.
After about 10 minutes of gassing a dirty-white precipitate forms, which consists of ephedrine hydrochloride + the hydrochlorides of the other ephedra alcaloids. The gassing is then stopped, all tubing connections are taken apart and the raw alcaloid hydrochlorides are filtered (buechner funnel). The raw alcaloid salts are mixed with 30 ml acetone in a 250 ml beaker and heated to boil on a water bath. About 10 ml of methanol are then added, the salts must have gone totally into solution. A spatula tip full of activated carbon is added (carefully, the solution might spontaneously foam up and boil over), the solution is brought shortly to full boil and is then filtrated hot.
From the hot filtrate (in my experience it is best to bring the filtrate again to a boil and then take it from the water bath) the ephedrine hydrochloride is precipitated by adding acetone until crystals begin to form and letting the mixture cool to room temperature. After 6 hours of standing the precipitated ephedrine hydrochloride is filtered off, washed with a small volume of acetone and dried.
If the product shows impurities (testing is done by TLC), it is recrystallized by dissolving in 15 times its volume of methanol, adding about 100 times its volume of acetone and letting the mixture sit in the refrigerator for 6 hours at 4 °C. The then precipitated crytals are filter off by means of a sinter glass filter. After one or two recrystallizations no more impurities are shown by TLC.
Yield: varies from 0.5 g to 1.5 of ephedrine hydrochloride
TLC: silca gel 60 F(254 nm), eluent consists of 49 parts ethanol, 49 parts dichloromethane, 2 parts ammonia solution 25 % (all parts by volume)
[Translated from:
Stahl/Schild: "Isolierung und Charakterisierung von Naturstoffen"; 1986, Fischer Verlag.
Italics mine]
[edit: The ephedra species used was ephedra sinica STAPF]
Quidquid agis, prudenter agas et respice finem!